Abstract Prostate cancer (PC) is one of the most frequent tumors in men. Despite recent progress, the disease is still incurable once resistance to castration therapy occurs. Tumor progression is strongly mediated by altered molecular exchanges between cancer cells and the surrounding milieu that originate at the primary sites. However, the mechanisms regulating the response of the stroma to the tumor, which ultimately promote PC progression are still largely unknown. Our laboratory discovered a new type of tumor-derived extracellular vesicle (EV), which are referred to as ?large oncosomes? (LO), can harbor more abundant molecular cargo that is distinct and more potently bioactive than that carried by exosomes. The rationale for this proposal derives from our preliminary observations in patients that LO abundance in the circulation correlates with PC progression. Our functional data demonstrate that LO can activate oncogenic signaling in fibroblasts, which respond to LO uptake by activating MYC and SPI1 and by induce a transcriptional program that promotes angiogenesis and stimulates tumor growth. The overarching goal of this project is to determine the functional role of LO in PC progression and metastasis. We hypothesize that LO functionally reprogram normal prostate-associated fibroblasts (NAF) toward a phenotype that is driven by MYC and SPI1 activation. These results strongly suggest that tumor-derived LO might activate intercellular responses that are specific to this subtype of extracellular vesicle. Our hypothesis will be tested with three Specific Aims: Aim 1: To investigate the role of LO-induced fibroblast activation in PC progression. Aim 2: To find evidence that the LO- induced transcriptional program is active in PC patients with clinically significant disease. Aim 3: To test if LO and/or Exo derived from PC patient and PDX specimens promote castration resistance and/or bone metastasis. We will use a combination of complementary in vitro and animal orthotopic models as well as focused approaches involving genome editing, molecular barcodes, and a Cre-Lox reporter in vivo system. Our study will determine if the transcriptional program induced by LO in vitro drives tumor progression and metastasis in vivo. Additionally we will determine if this transcriptional program can also be identified in patient specimens and if it indicative of tumor progression. Finally, our study will provide evidence for LO abilities to induce metastasis of indolent PC cells.